Color Management with ICC and DCP profiles on CUDA

Updated: August 27, 2026

Fastvideo color management applies ICC and DCP color profiles on the GPU so camera images reproduce accurate colors in real time. Profile-based color transforms run on CUDA inside the image pipeline, giving correct, consistent color without a CPU bottleneck.

How these numbers were measured. Fastvideo benchmarks are measured on NVIDIA GPUs, timing pure GPU processing at the resolution, bit depth, format and quality stated on this page. See the full benchmark methodology and test environment and the Fastvideo SDK benchmark report (PDF). Our tools are open source on GitHub and built on NVIDIA CUDA.

Check a single module. Every module in the Fastvideo SDK comes with a command-line sample. Run it on your own images on your own GPU and compare the result with the figures published here.

Check the whole pipeline. A module in the middle of a pipeline cannot be judged on its own: its input is hard to prepare, and its output is an intermediate product you cannot look at. What matters is how it changes the final image and the total time of the schema. Two of our applications run the complete pipeline and report the time of every stage — FastVCR (live cameras or RAW files from disk, so no camera is required) and Fast CinemaDNG (DNG, offline). The benchmark module is included in the free builds of both.

Detailed benchmark data. Measured results are published on our benchmark pages: all benchmarks.

Calculations of camera gamut gives opportunity to evaluate ability to reproduce right colors by the camera. This is the problem which happens quite often: if you need to replace color camera with a new one, you can't be sure that colors on a new camera will be the same as it used to be. Right solution for that task is camera calibration with ColorChecker to create its own ICC/DCP profile. The software should be able to have color management to reproduce right colors.

Standard workflow according to ICC includes color transform from Input Device Color Space (camera) to PCS (Profile Connection Space) and then transform from PCS to Output Device Color Space (monitor). PCS is the reference color space in which colors are encoded in order to provide an interface for connecting source and destination transforms. The PCS values constitute an encoding of a CIE colorimetric specification.

That Color Management software supports ICC v4 profile-based color management, including support for calibrated and profiled displays. Software engine doesn't use Windows Color System (WCS) which is available on Windows Vista and other operating systems.

We wrote these algorithms ourselves

Image processing has always been a trade-off between quality and speed. The methods that give a good result take minutes per frame; the fast ones leave halos and artifacts. The reason is the compute budget: on a CPU there was never enough of it, so algorithms were simplified — and much of the industry still runs on simplifications accepted decades ago.

Fastvideo’s codecs and ISP modules are neither ports of that legacy nor wrappers around third-party code. We designed — and in many cases invented — the algorithms behind our demosaicing, denoising, tone mapping and JPEG/JPEG2000 codecs, and wrote them for CUDA from the start, on top of image-processing expertise built since 2009. An algorithm moved to the GPU still carries the compromises it was simplified for; ours were never made under that constraint. On a modern GPU the budget is large enough that, for practical workloads, the choice between quality and speed no longer has to be made.

That is why you get a complete, tuned imaging solution rather than a bag of GPU primitives — and why, when your pipeline needs something specific, we can change the algorithm itself, not just call it.

Color management workflow for camera applications

How does ICC color management work on the GPU?

Monitor color calibration with i1 Photo (X-Rite) spectrophotometer to create ICM-profile
Camera color calibration with ColorChecker to create ICC-profile
Parsing of camera and monitor profiles
Extraction of all necessary data from icc/icm-profiles
Color transform with input profile (PCSLAB camera profile)
- Transform with Input curves
- Transform with CLUT (transform from linear RGB to Lab space)
- Transform with Output curves
- Transform from Lab to XYZ profile connection space
Color transform from XYZ to output profile (PCSXYZ monitor profile)
- Reverse matrix transform from XYZ to RGB
- Reverse linearization transform with reverse TRC
- Image output to monitor via OpenGL
Color transform from XYZ to sRGB output profile (standard sRGB profile)
- Reverse matrix transform from XYZ to sRGB
- Reverse linearization transform with reverse TRC
- Save image to HDD (optional compression to JPEG or to JPEG2000)

Important

Output profiles of monitor and sRGB are different and it means that we can't save on HDD and show on the screen the same data. In most cases, when we apply monitor and sRGB profiles to the same input data, we get different results. It's really easy to check whether this is true or not. You can open JPEG image with GIMP or other software with color management support and choose Edit/Parameters/Color Management menu and then apply sRGB or monitor profile. If you have done monitor color calibration and got its profile, you will clearly see the difference.

About the author

Fyodor Serzhenko, PhD, is the founder and CEO of Fastvideo. He earned his PhD at the Moscow Institute of Physics and Technology (MIPT) in 1993. Since 2009 he has led the development of Fastvideo’s GPU-accelerated image codecs and ISP modules, including the Fastvideo SDK. Connect on LinkedIn.

Why you can trust these results

Fastvideo has built GPU-accelerated image processing software since 2009. Our codecs comply with the standards they implement — JPEG (ITU-T T.81 / ISO IEC 10918) and JPEG2000 (ITU-T T.800 / ISO IEC 15444) — and our tools are open source on GitHub. Every performance figure we publish is reproducible: see our benchmark methodology, download the Fastvideo SDK benchmark report (PDF), and measure it on your own GPU.